American Cinematographer - September 2004 - 100

important consideration. Secouet had
specified that he wanted the overall
image to have a 120° field of view, so
each

camera

had to have about

field of view. But for all the

a

30°

cameras

to

line up properly, they had to share a
nodal point, and the larger the camera,
the farther away the nodal point would
be. Moreover, each camera needed room
to move around and couldn't

weigh too
much; with every camera part multiplied

by eight, increases in weight added up
quickly.
Ultimately, Trageser concluded
that the best choice for the project would
be Sony DSR500 DVCam cameras. With
DVCam, the cameras could have 90minute loads, an important advantage
because the array
of

rigging and

required a great deal
alignment. Unlike

consumer-level cameras, the DSR500s
could be gen-locked together so that
their time codes would match, and their
shutters could be slaved to avoid the

possibility of having strobe lines in
image and not in another.
Based

on

one

his theories about the

rig, Trageser began drawing up plans to
scale, which helped him calculate how
big the array would have to be: 6' wide
by 41/2' tall. The cinematographer then
showed the designs to his key grip,
Michael Gambino, who thought the
concept might not be beefy enough. The
problem, Gambino explained, was that
Trageser planned on using speedrail.
Because speedrail is a combination of a
pipe, a joint and a nut, every time it's
tightened up there is some slip and give,
making it unsuitable for a rig that would
bear so much weight and take a pound¬
ing as it was driven along rough roads.
While Trageser and Gambino
were working out the construction
details for the rig, they consulted with

own

generator.
As

Trageser and Gambino were
talking in Nashold's shop, they saw one
of his employees TIG-welding a metal
frame together and realized that this
could be what they needed for the array.
They told Nashold what they needed for
the rig and asked him whether he could
do the job. Nashold responded, "You
draw it up, and I'll build it."
Trageser says Nashold brought
even more value to the project because
his years of sending out camera cars had
taught him about the stresses a road can
cause.

He

was

therefore able to enhance

Trageser's design to make it more stable
and secure. "They took my idea, refined
it and then built it," says Trageser. The
final rig was made out of 21/2n-diameter
aluminum square tubing that was TIGwelded together, with the two bases
made in an A-frame shape for more
stability.
Once the design was set,
Trageser and his crew began examining
how best to make the rig pan - by two
people moving it manually, or by motor¬
ization. A first attempt at motorization
using two 5" servo motors proved unsuc¬
cessful, as did a second try with motion
dampers. (When fully assembled, the rig
ended up weighing about 400 pounds,
and neither solution was robust enough
to deal with that much weight.) Finally,
they found a McMaster-Carr super¬
heavy-duty turntable that could smoothly
bear 150,000 pounds.
Meanwhile, Trageser's camera
assistant, Dean Gunderson, had been

working

on

the design of the video¬

control station, which included a cluster
of electronic equipment for every
camera. The chances for mechanical and
electrical failure

were enormous:

"Every

Lee Nashold at Camera Cars Unlimited

cable had to connect, and every connec¬
tor could possibly break," explains

to determine what kind of vehicle

Trageser. After

should to
was

use

to carry

they
it. The best option

determined to be

a

converted

Hummer, which could carry the rig on its
front and still get to the off-road loca¬
tions the filmmakers wanted to
addition to

use.

In

having plenty of room for
video-engineering area on
the back, the Hummer could carry its
crew

and

a

few instances of what
as "teething pains,"
everything was assembled, working and
weatherized against the rain and snow
of February. "All of that effort was in the
service of making the rig and the
Hummer into one big camera on one big
dolly," notes Trageser.
Trageser and Secouet could now
a

Secouet referred to

turn to their

goal: the creation of a
moving photo collage. Notes the cine¬
matographer, "Once you get the cameras
aligned, there's still some slop because
the nodal points aren't exactly the same.
Therein lies the interesting aspect of this
undertaking: the images aren't a perfect
match. If each
different falloff

one
or

has

flares

a

The

camera

rig

was

mounted

onto

a

converted

Hummer.

little bit of

differently, that's

all for the best." He adds that the frac¬
tured

viewpoint, with a vista shown in
eight perspectives instead of just one,
brought many nice surprises: "When you
go around a corner with a 120-degree
field of view, if you've got the array
panned to the right, the left two cameras
are almost looking down the road while
the right two are almost looking at the
side. When Yann started playing with the
zoom controls, it really started getting
interesting."
When first set up, the cameras
were aligned to a particular perspective,
with the horizon line at the dead center.
As the array was tilted, a weird skewing
would occur: the center would always
hold

fairly true, but the because the outer
American

Cinematographer 87
American Cinematographer - September 2004

Table of Contents for the Digital Edition of American Cinematographer - September 2004
